The Cognitive Neuropsychology of Delusions

The Cognitive Neuropsychology of Delusions
ROBYN LANGDON AND MAX COLTHEART

Abstract: After reviewing factors implicated in the generation of delusional beliefs,
we conclude that whilst a perceptual aberration coupled with a particular type of attri-
butional bias may be necessary to explain the specific thematic content of a bizarre
delusion, neither of these factors, whether in isolation or in combination, is sufficient
to explain the presence of delusional beliefs. In contrast to bias models (theories which
explain delusion formation in terms of extremes of normal reasoning biases), we advo-
cate a deficit model of delusion formation—that is, delusions arise when the normal
cognitive system which people use to generate, evaluate, and then adopt beliefs is
damaged. Mere bias we think inadequate to explain bizarre delusions which defy com-
monsense and persist despite overwhelming rational counter-argument. In particular,
we propose that two deficits must be present in the normal cognitive system to explain
bizarre delusions: (1) there must be some damage to sensory and/or attentional-orient-
ing mechanisms which causes an aberrant perception—this explains the bizarre content
of the causal hypothesis generated to explain what is happening; and (2) there must
also be a failure of normal belief evaluation—this explains why a hypothesis, implausible
in the light of general commonsense, is adopted as belief. This latter deficit occurs, we
suggest, when an individual is incapable of suspending the natural favoured status of
direct first-person evidence in order to critically evaluate hypotheses, given equal pri-
ority whether based on direct or indirect sources of information. In contrast, delusions
with ‘ordinary’ content may arise when a single deficit of normal belief evaluation
occurs in the context of an extreme (but normal) attentional bias, thus causing failure
to critically evaluate hypotheses based on misperceptions and misintrepretations of
ambiguous (but ordinary) first-person experience.
1. Introduction
Delusions are a key feature of psychosis. Together with hallucinations—the
other so-called ‘reality distortion’ symptom—these symptoms mark what we
commonly conceive of as ‘madness’. In its Diagnostic and Statistical Manual of
Mental Disorders, Fourth Edition (DSM-IV), the American Psychiatric Associ-
ation (1994) defines a delusion as:
A false belief based on incorrect inference about external reality that

is firmly sustained despite what almost everyone else believes and
We would like to thank participants in the workshop held at Macquarie University in July/August
1998 for helpful discussion of many issues discussed in this paper, especially Andy Young. Special
thanks must also go to Martin Davies and Greg Currie for invaluable advice on an earlier draft
of this paper.
Address for correspondence: Robyn Langdon, Macquarie Centre for Cognitive Science, Mac-
quarie University, Sydney, NSW 2109, Australia.
Email: robynKrosella.bhs.mq.edu.au.
Mind & Language, Vol. 15 No. 1 February 2000, pp. 184–218.

 Blackwell Publishers Ltd. 2000, 108 Cowley Road, Oxford, OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA.
The Cognitive Neuropsychology of Delusions 185
despite what constitutes incontrovertible and obvious proof or evi-

dence to the contrary. The belief is not one ordinarily accepted by
other members of the person’s culture or subculture (e.g. it is not an
article of religious faith) (p. 765).
Despite concern about how such a definition should be operationalized by

clinicians, and despite debate about whether delusions should be distinguished
from normal beliefs in terms of dichotomy or continuum (see, for example,
Garety and Hemsley, 1994), there is some consensus that the characteristic
features of delusions are:
I that they run counter to the beliefs held by others within the same
socio-cultural environment;
I that they defy rational counter-argument;
I that they are maintained despite overwhelming counter-evidence.
Deluded individuals can also be found to espouse their beliefs with a note of
absolute conviction—as if the delusional beliefs are self-evident incontrovert-
ible truths—despite the fact that others clearly find the deluded individuals’
ideas to be patently absurd (Cutting, 1985; Sims, 1988).
Delusions come in a myriad of forms. In psychiatric textbooks and in clini-
cal rating scales, delusions are usually classified by thematic content; for
example, there are bizarre delusions, grandiose delusions, persecutory
delusions, and somatic delusions. Delusions can also vary in scope: some
deluded individuals present with tightly circumscribed delusions, whereas other
individuals present with a propensity to be delusional about anything that grabs
attention. For example, Stone and Young (1997) have noted the contrast
between monothematic delusions that can arise after right-hemisphere brain
damage—such as the belief that your arm is someone else’s (Halligan, Marshall
and Wade, 1995) or the belief that people in disguise are following you (de
Pauw, Szulecka and Poltock, 1987)—and some cases of schizophrenia where
the deluded individual inhabits a separate ‘reality’ peopled by imaginary beings
and filled with delusional events. Here the delusions are polythematic.
The aim of this paper is to review research investigating the factors that
may contribute towards delusion formation in order to develop a theory of
the processes which underpin normal belief generation and belief evaluation;
a theory which can then be used to explain delusions of all types. As such,
our aim here is to advance the field of cognitive neuropsychiatry by applying
the logic of cognitive neuropsychology in order to explain delusions—a symp-
tom which is generally described as psychiatric. The approach of cognitive
neuropsychology has two main aims (Coltheart, 1984; Ellis and Young, 1988).
The first is to explain aberrant behaviours and symptoms seen in patients after
brain damage in terms of what has been lost (damaged or disconnected) and
what remains intact in a model (or theory) of normal cognitive functioning.
 Blackwell Publishers Ltd. 2000
186 R. Langdon and M. Coltheart
The second is to evaluate models of normal cognition in terms of how well

they explain the various patterns of spared and dysfunctional cognitive
capacities observed in neurological patients. Since our aim here is to develop
a model of the normal cognitive system of belief generation and belief evalu-
ation which can then explain all types of delusions in terms of damage to one
or more components of that cognitive system, we will use as our primary
examples cases of ‘organic’ delusions and cases of ‘psychotic’ delusions. The
former refer to delusions which occur after known organic brain damage and
the latter refer to delusions found in psychiatric disorders where the nature of
the neuropathology remains unspecified. Historically, this distinction has also
been described in terms of the ‘organic’ versus ‘functional’ psychoses (Frith,
1992). In particular, we will focus on cases of misidentification delusion which
can occur after known right-hemisphere brain damage (primarily, Capgras
delusion—the belief that an impostor has replaced a loved one) and various
types of psychotic delusions found in cases of schizophrenia. The contrast
between ‘organic’ misidentification delusions and ‘psychotic’ schizophrenic
delusions also highlights another relevant contrast in the study of delusions,
and that is the distinction between circumscribed monothematic delusions and
widespread polythematic delusions, as noted earlier. That is, whereas Capgras
patients will often present with a tightly circumscribed delusion—they may
only be delusional about the impostor who is impersonating their loved one—
patients with schizophrenia will often present with a florid widespread
delusional system. However, it is worth noting here that this need not be the
case—that is, the circumscribed versus polythematic distinction does not always
map neatly onto the organic versus psychotic distinction. Some Capgras
patients do build a more extensive network of delusional beliefs stemming from
that first deluded belief about an impostor, and some schizophrenic patients do
present with tightly circumscribed delusions—for example, there are patients
with schizophrenia who have auditory hallucinations, specifically of the voices
commenting1 type, and who only develop delusional beliefs about the source
of their auditory hallucinatory experiences.
A useful distinction to bear in mind when considering the factors which
may contribute towards delusion formation is that between factors which play
a role in shaping the thematic content of different delusional beliefs and factors
which explain the presence of a delusional belief. In other words, there is a
difference between explaining what it is that a person comes to be deluded
about and explaining why that person is deluded in their thinking about
those ideas.
We begin by reviewing research investigating the role that perceptual aber-
1
Auditory hallucinations of the voices commenting type are experiences where patients hear
another voice commenting on their behaviour in the second person, as distinct from third-
person auditory hallucinations where patients hear two or more voices discussing them in
the third person, and auditory hallucinations where patients hear non-vocal sounds.
rations and attributional biases may play in the explanation of delusions. Our
line of argument will be to question whether either of these factors or both
in combination is sufficient to explain the presence of a delusional belief. We
will conclude that, whilst the presence of a perceptual aberration when coupled
with a particular type of attributional bias may be necessary to explain the
thematic content of a particular bizarre delusion, neither of these factors,
whether in isolation or in combination, is sufficient to explain the presence of
a delusional belief. We then consider research investigating whether deluded
individuals show deficits of reasoning in order to derive a model of normal
belief generation and belief evaluation which we then use to explain a variety
of delusions.
2. The Role of Perceptual Aberrations

One prominent and influential approach to explaining delusion formation has
come from Maher and colleagues (Maher, 1974, 1988, 1992; Maher and Ross,
1984; Maher and Spitzer, 1993) who proposed that delusions are meaningful
hypotheses which have been generated by normal reasoning processes to explain
aberrant perceptual experiences, such as hallucinations. In other words, Maher
takes the view that the presence of an aberrant perceptual experience is neces-
sary and sufficient to explain the presence of a delusional belief. Whereas
Maher was primarily interested in understanding the role of perceptual aber-
rations in explaining the generation of schizophrenic symptoms, Young and
colleagues (Ellis and Young, 1990; Wright, Young and Hellawell, 1993;
Young, 1994; Young, Leafhead and Szulecka, 1994; Young, Reid, Wright
and Hellawell, 1993) have, more recently, been interested in investigating the
possible role that perceptual aberrations may play in the generation of mis-
identification delusions, in particular Capgras delusion—the belief that a loved
one has been replaced by an impostor. In this section, we review empirical
research investigating the nature of perceptual aberrations implicated in both
Capgras delusion and delusions of alien control, the latter being among the
first-rank symptoms of schizophrenia.
Individuals with Capgras delusion believe that a loved one, usually their
spouse, has been replaced by an impostor who looks identical to the real loved
one. Researchers investigating the factors that may be implicated in the gener-
ation of Capgras delusion have proposed that this delusional belief arises when
an individual loses the sense of affective responsiveness which normally co-
occurs with explicit recognition of a familiar face (Ellis and Young, 1990).
The resultant discordance between ‘he looks right’ and ‘he doesn’t feel right’
is then resolved by the individual adopting the belief that the loved one has
been replaced by an impostor—an idea which, plausible or no, does have
explanatory sufficiency, given the nature of the perceptual aberration. The
suggestion that lessened autonomic responsiveness to a loved one’s face does
have some role to play in the generation of Capgras delusion has been investi-
gated by Ellis, Young, Quayle and de Pauw (1997a). In their study, Ellis et
al. (1997a) predicted that, if people with Capgras delusion are under-responsive
to familiar faces, then they should fail to show a pattern of autonomic discrimi-
nation2 between familiar and unfamiliar faces when compared to both non-
clinical control subjects and psychiatric patients who were taking similar anti-
psychotic medications but who were experiencing different types of delusions
(e.g. delusions of persecution). Furthermore, these researchers predicted that,
since Capgras patients are deluded about a loved one’s face but may be able
to recognize that loved one’s voice when encountered on the telephone (see
Ellis, Young, Quayle and de Pauw, 1997b), the autonomic under-responsive-
ness observed in these patients should be circumscribed—that is, specific to
face recognition. Both hypotheses were supported by the data. Familiar and
unfamiliar faces produced equal degrees of affective response in Capgras
patients, in contrast to both control groups who showed significantly greater
autonomic responsiveness to familiar faces. Furthermore, this was not due to
some general under-responsiveness of the Capgras patients, since these patients
showed normal autonomic orienting to a novel auditory tone and a normal
pattern of autonomic habituation to repetitions of that tone.
In the case of delusions of alien control in schizophrenia, patients who
experience these delusions express the belief that their actions are being con-
trolled by an external agent—sometimes an alien being, sometimes a spiritual
power. Note here that a delusion of alien control is not the same thing as
believing that one’s chosen course of action has been strongly influenced by
the views of significant others, as is the case, for example, in missionaries who
believe that they are acting out the word of God. Instead, patients with
delusions of alien control describe experiences of disembodied or deper-
sonalized agency—for them, there is no sense of personal choice. They are
performing the actions, but someone else is pulling the strings. Researchers
investigating the factors which may contribute towards the generation of
delusions of alien control have proposed that these patients experience a discor-
dant sense of passivity which arises when normal on-line self-monitoring is
disrupted, resulting in loss of the sense of self-generation which normally co-
occurs with self-initiated action (Frith and Done, 1986, 1989; Frith, 1992). In
more detail, the idea here is that normal self-initiation of action involves the
coordination of two matching signals: (1) a motor instruction to initiate bodily
movement; and (2) a record of that instruction (a reafference copy of the
intended act) which is registered with a central monitor. In the case of
delusions of alien control, it is argued, registration of this second signal is
disrupted. Consequently, these individuals see and feel themselves acting in
2
In this study, skin conductance responses (SCRs) were recorded while subjects were shown
a series of predominantly unfamiliar faces, with occasional familiar (famous) faces interspersed.
The difference in mean SCR to familiar and unfamiliar faces was then used to index auto-
nomic discrimination of familiar and unfamiliar faces.
ways which are appropriate and meaningful in the circumstances in which they
find themselves, but with no associated sense of self-agency—from which they
conclude that they are robots or puppets being manipulated by an external
power.
One way of testing the hypothesis that defective self-monitoring plays a
role in the generation of delusions of alien control derives from the idea that
central monitoring underpins our normal ability to make rapid corrections of
inappropriate responses before needing to wait to see the consequences of an
error (Angel, 1976; Frith and Done, 1986). That is, by on-line monitoring an
intended act (via monitoring of a reafference copy) and matching that against
the overall goal, we can quickly identify errors and sometimes short-circuit
mistakes before they are fully acted out, as when interrupting speech mid-
word in order to correct an error (Leudar, Thomas and Johnston, 1994). Frith
(1992) describes the capacity as follows:
we can be aware that an intended response was wrong after we have

initiated that response, but before the consequences of that response
are visible (p. 83).
If we accept such a model of normal action control which involves two

processes, one being based on internal monitoring of intentions and the other
being based on external monitoring of the consequences of action (Frith and
Done, 1986), and if it is the case that patients with delusions of alien control
do have defective internal monitoring, then these patients should perform
poorly on error-prone motor response tasks when access to visual feedback is
denied. Frith and Done (1989) have tested this hypothesis by investigating
patterns of error correction in schizophrenic patients with alien control experi-
ences,3 schizophrenic patients without such experiences, non-schizophrenic
psychiatric controls and normal controls. In their study, presented as a video
game, subjects moved a joystick in order to shoot a target which could appear
at different locations. The difficulty of making correct moves was manipulated
in order to induce a high error rate. In one version of the task, subjects moved
their joystick and saw the trajectory of their shot being tracked across the
computer screen (a shot took 2800 msecs to reach a target). Under these cir-
cumstances, all subject groups corrected 100% of their errors within
2000 msecs. In the second version of the task, the trajectory of a shot was
hidden (behind a wall depicted on the screen) for the first 2000 msecs after
making a response. Under these circumstances—with no visual feedback as to
whether a mistake had been made for the first 2000 msecs—only 8% of patients
3
Alien control experiences, as operationalized by Frith and colleagues (Frith, 1992; Frith and
Done, 1989), include delusions of alien control and also auditory hallucinations of the voices
commenting type which are interpreted as a patient ‘hearing’ his or her internal commentary,
disconnected from any sense of self-generation.
with alien control experiences corrected their errors within 2000 msecs, com-
pared to 80% of schizophrenic patients without alien control experiences, 40%
of psychiatric controls and 67% of normal controls.
However, although empirical findings such as those cited above clearly sup-
port the view that perceptual aberrations play some role in the generation of
bizarre delusions, such as Capgras delusion and delusions of alien control, it
cannot simply be the case that a perceptual aberration is sufficient to explain
the presence of a delusional belief. Consider, for example, why it is that the
deluded Capgras patient and the schizophrenic patient with delusions of alien
control fail to accept the perfectly plausible hypothesis that brain damage or
dysfunctional neurochemistry has caused their senses to provide unreliable
information about the environment or themselves. Even more puzzling, why
do these individuals fail to reach the simple conclusion that ‘something strange
must be going on, but I don’t know what’ (Stone and Young, 1997)?
If the presence of a perceptual aberration were a sufficient condition for
the presence of a delusional belief, then any individual who experienced an
aberrant perceptual experience should develop a delusional belief. But that is
not so. Take, for example, the case of phantom limb experiences. Many indi-
viduals who have had a limb amputated have the vivid experience that their
lost limb is still present, and in many cases also painful. Recent studies using
microelectrode recording, microstimulation, and various imaging techniques,
have provided evidence that part of the thalamus which originally responded
to nerve inputs from a limb which has been amputated can remain functional
and can mislocalize sensory stimulation from other sources after cortical
reorganization of connections to the thalamic ‘limb’ neurons (Davis et al.,
1998; Flor et al., 1998; Ramachandran and Hirstein, 1998).
The critical point here is that although something is dysfunctional in the
brains of amputees with phantom limb experiences, and although these indi-
viduals have no direct access to the nature of the dysfunction which causes
their aberrant experience—feeling a limb which cannot be seen and which
doctors say has been removed—these individuals do not become delusional.
They do not develop the belief that their arm is still intact and that ‘evil’
doctors have made it invisible; instead they readily accept that something has
gone wrong (in their brain) which makes it feel to them as if their limb is still
present, when in fact they know that their limb has been removed.
Even more compelling grounds for rejecting the view that the presence of
a perceptual aberration is sufficient to explain the presence of a delusional
belief comes from evidence that some non-deluded individuals may be experi-
encing the same, or at least very similar, perceptual anomalies to those found
in patients with Capgras delusion and delusions of alien control. To begin,
recall that, in the case of Capgras delusion, it was argued that intact explicit
recognition of a loved one’s face coupled with the loss of affective response
to that loved one’s face generates a discordant perceptual experience which is
then resolved by the individual adopting the belief that the loved one has been
replaced by an impostor. Empirical support for that view came from evidence
that Capgras patients fail to show a normal pattern of autonomic discrimination
between familiar and unfamiliar faces. However, Tranel, Damasio and Damasio
(1995) have found that patients with damage to bilateral ventromedial frontal
regions of the brain also fail to autonomically discriminate between familiar
faces (which they recognize explicitly) and unfamiliar faces; yet these patients
were not delusional. If we assume that the dissociation between intact explicit
recognition of a familiar face and the loss of autonomic responsiveness to that
face generates the same perceptual aberration in both Capgras patients and
Tranel et al.’s (1995) frontal patients, then what makes for the formation of a
delusional belief in Capgras patients or, instead, what is it that protects the
frontal patients from becoming delusional?
But perhaps assuming that a similar pattern of empirical data indexes a simi-
lar perceptual anomaly in Tranel at al.’s (1995) frontal patients and Ellis et al.’s
(1997a) Capgras patients is unjustified. It is possible that inability to generate
discriminatory autonomic responses to familiar and unfamiliar faces occurs for
different reasons in non-deluded patients with bilateral ventromedial frontal
lesions and Capgras patients, some of whom are known to have different sites
of brain damage—in particular, occipito-temporal or temporal-parietal lesions
(Stone and Young, 1997). And perhaps these different reasons lead to qualitat-
ively distinct phenomenal experiences. It is worth noting, in this context, that
the Tranel et al. (1995) study and the Ellis et al. (1997a) study differ in that
the former did not demonstrate that the under-responsiveness of their patients
was circumscribed. Indeed, Damasio, Tranel and Damasio (1991) have
reported elsewhere that patients with bilateral ventromedial frontal lesions fail
to show normal autonomic responses to emotionally charged visual stimuli
such as pictures of mutilations and social disasters. It is therefore possible, that,
even though Tranel et al.’s (1995) frontal patients show empirical evidence
of a discordant mismatch between intact explicit face recognition and loss of
autonomic face recognition, if that mismatch occurs in the context of general
affective flatness (which may well be the case in these patients), then the result-
ant perceptual experience (when encountering a familiar face) may differ from
that of Capgras patients and may indeed not even register as aberrant.
Let us then consider two distinct clinical disorders where patients present
with very similar phenomenological self-reports of their experiences. The first
of these is depersonalization disorder, classified in DSM-IV as a dissociative
disorder which is characterized by altered perceptions and experiences of self.
These individuals describe their experiences in very much the same way that
schizophrenic patients with delusions of alien control do—that is, they describe
experiences of being outside of their bodies, of feeling like mechanical robots,
and of having no sense of being in control of their own actions (Davison and
Neale, 1998). Indeed, the similarity is so striking that patients with deper-
sonalization disorder have been misdiagnosed as having schizophrenia. How-
ever, there is a critical feature which distinguishes the two clinical disorders:
whilst the insightful patient with depersonalization disorder describes his or

her experience in ‘as if’ terms (it is as if an alien is controlling my actions),
the deluded schizophrenic patient with passivity experiences lives the concrete
reality (an alien is controlling my actions). In other words, what distinguishes
these two disorders is precisely whether or not some critical faculty which
underpins normal reality testing, or belief evaluation, is operating.
Sno (1994) has presented a somewhat similar argument when contrasting
non-pathological déjà vu experiences and some cases of reduplicative param-
nesia. Non-pathological déjà vu experiences occur when an individual experi-
ences an inappropriate subjective impression of familiarity about the present
situation (as if one’s present surroundings are identical to some unspecified
place that one has visited in the past), whilst at the same time being aware
that this is an impossibility. In contrast, reduplicative paramnesia is the deluded
belief that a familiar place (or person) from the past has been duplicated in the
present (Breen, Caine, Coltheart, Hendy and Roberts, this volume; Hakim,
Verma and Greiffenstein, 1988). To illustrate, reduplicative paramnesia often
takes the form of patients believing that they are in a hospital which they have
been in before, usually in their home-town, when in fact the hospital in which
they are currently a patient is located in a completely different city; a belief
which is persistently sustained despite compelling counter-evidence (Benson,
Gardiner and Meadows, 1976; Filley and Jarvis, 1987; Roehrenbach and
Landis, 1995). The implication here is that both non-deluded individuals with
déjà vu experiences and deluded patients with some forms of reduplicative
paramnesia may be experiencing a common perceptual aberration (most likely
caused by inappropriate activation of memorial familiarity) which is then inter-
preted differentially depending upon the presence or absence of some critical
faculty involved in normal belief evaluation.
In sum, the empirical findings reviewed in this section support the view
that, although perceptual aberrations contribute towards the generation of
bizarre delusions, such as Capgras delusion and delusions of alien control, the
presence of a perceptual aberration is not sufficient to explain the presence of
a delusional belief, since there exist individuals who experience very bizarre
perceptual experiences—for example, the feeling that an amputated limb is
still present and indeed painful—who are not delusional. Furthermore, there
are grounds for thinking that the same perceptual aberration can be found in
both non-deluded individuals and deluded patients.
In the following section, we consider whether the presence of a perceptual
aberration when coupled with a particular type of attributional bias, or a stat-
istically deviant attributional bias, is sufficient to explain the presence of a
delusional belief.
3. The Role of Attributional Biases
Attribution theory is that area of social psychology concerned with understand-
ing how ordinary people explain why things happen (Hewstone, 1983, 1989).
One of the historical forerunners of attribution theory is the ‘naı̈ve psychology’

of Heider (1944, 1958) who, for example, conceptualized potential causal fac-
tors as being either internal to an actor (ability, effort) or external (task-dif-
ficulty, luck). In terms of that dichotomy, ‘person’ attributions were seen to
be more likely than ‘situational’ attributions since a person, it was argued,
fits the prototype of causal origin—an attributional bias which can have nasty
consequences since a particular person in a group may be punished for some-
thing bad that has happened in order for the rest of the group to effect some
sense of control over frightening circumstances. Another important contri-
bution from Heider is the idea that perceived similarity between an ‘actor’
and an ‘act’ can determine the likelihood of a causal attribution. For example,
a ‘bad’ action is more likely to be attributed to a ‘bad’ person than a ‘good’
person. Now there are plenty of examples of this type of attributional bias
operating—just think about our jail populations and how ‘badness’, when
defined along racial lines, might be operating here.
In more recent times, Weiner (1988, 1990) has had a major impact on
attribution theory by extending the properties, or dimensions, of cause beyond
internal/external locus. According to Weiner, causal biases can be theoretically
classified along three dimensions: locus (internal/external); stability
(stable/unstable); and controllability (controllable/uncontrollable). For other
researchers, the degree of globalness is also seen to be an important dimension
of causality (Seligman et al., 1979). Much of the present-day research
investigating the influence of attributional biases uses self-report questionnaires
which incorporate these various dimensions of causality. For example, in the
Attributional Style Questionnaire (ASQ) (Peterson et al., 1982; Seligman et
al., 1979), subjects are asked to generate likely causes for six positive events,
and six negative events, and they are then asked to self-rate those causes on
dimensions of internality (that is, the degree to which events are attributed to
self or to external causes), stability (that is, whether causes will persist into the
future), and globalness (that is, the degree to which causes can influence a
range of events beyond the original event being considered). To illustrate,
people might be asked to imagine the positive event of meeting a friend who
then compliments them on their appearance, or they might be asked to
imagine the negative event of looking for a job and not being able to find
one. For each imagined positive and negative event, subjects must then write
down the one major cause for that event and they must then make three
separate ratings on a linear scale: (1) how much do they think that cause is
due to other people/circumstances or due to themselves; (2) whether they
think that cause will always be present or is unlikely to happen again; and (3)
whether that cause influences just this particular situation or influences all situ-
ations in their life.
In clinical contexts, the ASQ has been used traditionally with depressive
patients in order to reformulate the learned helplessness model of depression
in terms of personality dispositions (Abramson, Seligman and Teasdale, 1978).
For example, Sweeney, Anderson and Bailey (1986) have conducted a meta-
analysis of studies investigating attributional style in depressed individuals and
found that as attributions for negative events became more internal, stable and
global, depression increased. More recently, Bentall and colleagues (Bentall,
Kaney and Dewey, 1991; Kaney and Bentall, 1989; Lyon, Kaney and Bentall,
1994) have adopted attribution theory in order to investigate whether a parti-
cular type of attributional bias is implicated in the generation of delusional
beliefs. For example, Kaney and Bentall (1989) first used the ASQ with
delusional paranoid patients, depressive patients and normal controls and found
that both deluded patients and depressed patients were more likely to make
excessive global and stable attributions for negative events. Of more interest,
however, was evidence that patients with persecutory delusions were more
likely to make excessive external attributions for negative events and excessive
internal attributions for positive events, when compared to normal controls—
the exact reverse of the pattern found in depressed patients. These results have
since been replicated by Candido and Romney (1990) and extended in various
other studies which have also investigated the role of self-esteem and self-
concept (Bentall et al., 1991; Bentall and Kaney, 1996; Kinderman et al., 1992;
Kinderman and Bentall, 1996a, 1996b, 1997; Lyon et al., 1994). Based on this
work, Bentall, Kinderman and Kaney (1994) have developed a general theory
of persecutory delusions wherein threat-related information in the environ-
ment activates self/ideal discrepancies which then promote a self-defensive
attributional bias towards locating the cause for negative events in external
sources. To try to put it more simply, the idea here is that deluded patients
with persecutory delusions try to maintain self-esteem by avoiding discrep-
ancies between their ‘ideal’ self (who they want to be) and their ‘actual’ self
(as indicated by information in the world) by externalizing the cause for nega-
tive events and internalizing the cause for positive events. For a more detailed
account of this position and a review of empirical research investigating the
role of self/ideal discrepancies in some deluded subjects, readers are referred
to Garety and Freeman (1999).
For the purposes of the present paper, we mention the Bentall model
because it raises the possibility that a perceptual aberration when coupled with
an attributional bias to externalize the cause for negative events—a bias which,
evidence suggests, is present in deluded paranoid patients—may be sufficient
to explain the presence of a delusional belief. At first sight, that sounds a
perfectly plausible notion. Consider, for example, the case of Capgras delusion.
It has been proposed that Capgras delusion arises when the loss of affective
responsiveness to a loved one’s face generates an aberrant, perhaps frightening,
negative experience (‘he looks right but doesn’t feel right’) which is then mis-
attributed to something going wrong in the world (an impostor has replaced
my loved one) rather than something going wrong internally (brain damage
has caused me to lose affective responsiveness to familiar faces) (Wright et al.,
1993; Young et al., 1993). Further support for this view comes from evidence
that Capgras patients often present with a pervasive mood of

suspiciousness/paranoia (Fleminger and Burns, 1993) which, as noted earlier,
is known to be associated with an externalizing bias for negative events.
However, not every deluded subject has an external bias for negative events.
Indeed, rather than one particular type of attributional bias—that is, an exter-
nalizing bias for negative events—being associated, in general, with the pres-
ence of delusional beliefs, it appears that different types of attributional bias
can account for different delusional misinterpretations of the same perceptual
aberration. For example, the two distinct delusions of Capgras delusion (the
belief that an impostor has replaced a loved one) and Cotard delusion (the
belief that you are dead) are said to arise when the same perceptual aberration
(lack of affective response to personally significant faces) interacts with different
attributional styles; a suspicious externalizing bias for negative events in the
case of Capgras delusion and a depressive internalizing bias for negative events
in the case of Cotard delusion (Young et al., 1992, 1994; Wright et al., 1993).
To clarify, whereas the suspicious Capgras patient resolves the discordant per-
ception—‘he looks right but doesn’t feel right’—by externalizing blame in the
world (an impostor is impersonating my loved one), the Cotard patient, who
typically presents in the context of a general depressive state, internalizes blame
and concludes that something must be drastically wrong with themselves and
not with others—hence they must be dead. Indeed, Wright et al. (1993) have
reported the case of a patient who shifted from having Cotard delusion to
Capgras delusion as his mood changed from depression to suspiciousness
and paranoia.
And it is not only among misidentification delusions that we can find
examples of the same perceptual aberration being delusionally misinterpreted
in different ways, depending upon an individual’s particular style of attri-
butional bias. Consider here schizophrenic patients who experience the sen-
sation that their thoughts are transparent to others. Let us for the moment
assume that this is a perceptual aberration in the same way that loss of the sense
of self-generation normally associated with self-initiated action is a perceptual
aberration. However, in this case the idea would be that there is loss of the
sense of subjective privacy associated with ‘hearing in mind’ one’s own
thoughts (David, 1994). We say more about this perceptual anomaly later. For
now, the point here is that this same perceptual aberration may give rise to
different types of ‘loss of boundary’ delusions, depending upon an individual’s
particular attributional bias. That is, for individuals who tend to internalize
blame for negative events, that aberrant perceptual experience may give rise
to thought broadcast delusions—that is, the belief that you are broadcasting
or transmitting your thoughts to others; whereas individuals who tend to exter-
nalize blame for negative events may develop mind-reading delusions or
thought insertion delusions—that is, the belief that others have the power to
read your mind or the belief that others can insert thoughts into your mind.
A perceptual aberration and a particular type of attributional bias—an exter-
nalizing bias for negative events—cannot be jointly sufficient to explain the

presence of a delusional belief since there are some deluded individuals who
have the opposite attributional bias—that is, an internalizing bias for negative
events. However, perhaps what is critical here is not the direction of the bias
but the degree of bias. All of the examples of delusions which we have
described thus far are consistent with the notion that a perceptual aberration
when coupled with a statistically deviant degree of attributional bias, in which-
ever direction—whether a tendency to internalize blame or externalize blame
for negative events—may be sufficient to explain not only the thematic content
of a particular delusion but also why that individual becomes deluded about
that content.4 However, even that cannot be so, since not all deluded individ-
uals show evidence of a statistically deviant degree of attributional bias. Here
we refer to work by Sharp, Fear and Healy (1997) who investigated the attri-
butional styles of paranoid and/or grandiose patients and non-paranoid/non-
grandiose patients with somatic delusions or delusions of morbid jealousy. In
their study, Sharp et al. (1997) found that paranoid/grandiose patients did make
excessive external attributions for negative events when compared to normal
controls—thus replicating the pattern of results found by Bentall and col-
leagues; however, these researchers also found that their non-paranoid/non-
grandiose patients, who were just as deluded as the paranoid/grandiose group,
did not differ statistically from normal controls in their tendency to internalize
or externalize the blame for negative events.
At this point, there is one other type of attributional bias which we will
mention before moving on to present an interim summary of the discussion
thus far. In contrast to individual attributional biases, we think it likely that
there is a specific attributional bias which all people have, and that is the
bias to favour personal-level causal explanations over subpersonal-level causal
explanations. The Capgras patient’s belief that an impostor has replaced a loved
one and the Cotard patient’s belief that he or she is dead are both examples
of personal-level causal explanations. That is, these beliefs locate cause some-
where in the interaction between self and the environment. In contrast, subp-
ersonal causal explanations locate cause in the subpersonal mechanisms which
underpin ‘my being me in the world’. The belief that dysfunctional neuroch-
emistry has caused me to fail to experience the feeling that I should have when
I see a loved one’s face is a subpersonal causal explanation. Our suggestion
here is that subpersonal causal explanations which may require us to question
the integrity of the subpersonal processes which provide us with information
about the world are intrinsically difficult for any of us to entertain. In other
words, it is part of the normal human condition, we think, to have an attri-
4
Up until now we have only discussed internalizing biases for negative events in connection
with depression. But having depression need not preclude being delusional—that is, delusions
can be found in cases of major depression.
butional bias which favours personal-level causal explanations over subper-

sonal-level causal explanations.
4. Interim Summary
We began this paper with a working definition of delusions: delusions are
beliefs which run counter to the general beliefs held by others within an indi-
vidual’s socio-cultural environment and which an individual maintains despite
rational counter-argument and overwhelming counter-evidence. Before
reviewing research investigating the factors which may contribute towards
delusion formation, we alerted readers to a useful distinction to bear in mind
when evaluating relevant empirical evidence. That is, there is a difference
between explaining delusional content—what it is that a person comes to be
deluded about; and explaining the presence of a delusional belief—why that
person is deluded in their thinking about those ideas.
Empirical evidence was reviewed which supported the existence of different
types of perceptual aberrations in different groups of deluded individuals, such
as those with Capgras delusion and those with delusions of alien control. Such
evidence leads us to conclude that the presence of a perceptual aberration is
necessary to explain why a delusion is bizarre, by which we mean having
aberrant thematic content which runs counter to the everyday commonsense
beliefs of others. The contrast here is with ordinary delusions—that is, delusions
which are about everyday things, such as believing that a loved one is having
an affair. We will say more about ordinary delusions later. For now, the point
to be noted here is that the presence of a perceptual aberration explains why
it is that such affected individuals generate bizarre delusions with aberrant con-
tent since, in these cases, the deluded individuals with perceptual aberrations
are not experiencing themselves or the world in the same way that normal
individuals are. It is therefore not surprising that they will generate bizarre
ideas which run counter to the everyday commonsense beliefs of others. Fur-
thermore, the existence of different types of perceptual aberrations (e.g. loss
of affective responsiveness to faces and loss of the sense of self-generation asso-
ciated with self-initiated action) helps us to explain why deluded individuals
can differ so widely with respect to the thematic content of their bizarre
delusions and may explain why deluded individuals differ so widely with
respect to the scope of their delusions. That is, some delusions may be mono-
thematic because only one perceptual aberration is present (e.g. the Capgras
delusion is solely about spouse replacement) whereas other delusions may be
polythematic because the deluded individual is bombarded with more than
one perceptual aberration (e.g. this may be the case in the classic example of
Dr Daniel Schreber, a German judge, whose highly complex delusional system
included beliefs about his being transformed into a female to be fertilized by
God and beliefs that doctors were committing soul-murder upon him
(Freud, 1963)).
However, the presence of a perceptual aberration, although necessary to

explain the content of a bizarre delusion, is not sufficient to explain the presence
of a delusional belief, since, as we have already stated, some individuals with
bizarre perceptual experiences do not have delusions and since there are
grounds for thinking that both non-deluded and deluded individuals can be
experiencing the same perceptual aberration.
We then considered the possibility that a perceptual aberration when
coupled with a particular type of attributional bias, or a statistically deviant
degree of bias, might be jointly sufficient to explain the presence of a delusional
belief. However, although an externalizing bias for negative events was found
to be associated with persecutory delusions, in general, and the Capgras
delusion, in particular, the presence of an externalizing bias for negative events
is not characteristic of all delusions. In particular, Cotard delusion, it was noted,
is said to arise when the loss of affective responsiveness to a loved one’s face
is delusionally misattributed to oneself being dead in the context of a dispo-
sition to internalize blame for negative events. Furthermore, not even the pres-
ence of a statistically deviant degree of bias—in whichever direction—is suf-
ficient to explain the presence of a delusional belief since groups of deluded
individuals can be found who do not differ significantly from normal subjects
in the extent to which they internalize or externalize the blame for negative
events.
In sum, perceptual aberrations may be necessary but are not sufficient to
explain the generation of bizarre delusions. Attributional biases, of any parti-
cular type or any particular strength, we think neither necessary nor sufficient
to explain the presence of a delusional belief. Deluded individuals are found
with all styles of attributional bias and without an abnormal degree of bias.
Furthermore, it is not pathological to have an attributional bias—whatever the
direction and whatever the strength. Differences in attributional bias play a
role in many non-clinical aspects of everyday life: determining drive for success
in the classroom, influencing ability to cope with stress, shaping responses to
marital conflict, and influencing worker attitudes in big organizations (Graham
and Folkes, 1990). We all differ to varying degrees in our disposition to allocate
blame internally or externally and we are all influenced by a universal bias to
favour personal-level causal explanations over subpersonal-level causal expla-
nations. The existence of such normal attributional biases can help us to explain
some of the individual variation in delusional content (e.g. Capgras delusion
versus Cotard delusion) and may help us to explain why deluded individuals
are not swayed by medical evidence of subpersonal causality; however, we still
need something more for a full causal account of delusions since non-deluded
individuals with phantom limb experiences and depersonalization disorder can
reason against their natural biases and accept a rational subpersonal explanation.
Why can’t deluded individuals? In the following section, we review research
investigating whether deluded individuals show deficits of reasoning and/or
logical thinking.
5. Studies of Reasoning in Deluded Individuals
Recall that Maher (1974, 1988, 1992) takes the view that a delusional belief
is a rational attempt to make sense of an aberrant perceptual experience. Now,
although we do not consider that the presence of a delusional belief is com-
pletely explained by the abnormality of a perceptual experience—for all of the
reasons listed above—there are some grounds for accepting Maher’s view that
reasoning processes in deluded individuals are not markedly abnormal.
Delusions do not require the presence of general intellectual deterioration.
Circumscribed delusions are found in non-demented individuals (Stone and
Young, 1997) and, despite evidence of poor performance on IQ tasks by
chronic schizophrenic patients, IQ deteriorates in only some schizophrenic
individuals, and that only after the onset of the disorder (Elliott and Sahakian,
1995). As for the possibility of other types of abnormal reasoning in deluded
individuals, much early interest in this issue focused more specifically on the
question of whether communicative deficits in schizophrenia are better under-
stood as abnormalities of communication or as abnormalities of thinking
(Kasinin, 1944). Some ideas proposed at that time were that schizophrenic
patients show poor abstract thinking (that is, patients are concrete thinkers
who interpret proverbs literally), poor concept formation (for example, on
object sorting tasks and categorisation tasks) and impaired syllogistic reasoning
(that is, patients are prone to make errors of the type ‘Diana is a princess, my
initial is D, therefore I am a princess’). Much of that research continues to
the present day, although the primary focus now is on executive function
deficits in schizophrenia. There is no scope to review this line of research in
any detail here. It will suffice to point out that concrete thinking and poor
abstraction can be found in non-deluded patients with frontal lesions or
depression. Furthermore, recent studies of syllogistic reasoning report little evi-
dence of differences between deluded patients and normal controls (Kemp et
al., 1997). So, although these abnormalities of thinking may be characteristic
of at least some schizophrenic patients, it is not likely that such abnormalities
of thinking will provide an explanation of delusions in general.
Overall, deluded patients appear to reason just as ably as normal subjects
on tests of logical deductive reasoning (conditional reasoning and syllogistic
reasoning) (Kemp et al., 1997). When differences are found, these appear to
be specific to inductive reasoning tasks involving probability judgements. For
example, Huq, Garety and Hemsley (1988) used an inferential reasoning task
in which subjects are shown two containers of coloured beads. One container
has pink and green beads in the proportion of 85 to 15, whereas the other
container has beads in the reverse proportion. The containers are placed out
of sight and the experimenter draws a sequence of beads from one container.
Subjects must indicate when they believe that they can make a judgement
about which container the beads are being drawn from and how confident
they are about their decision. Deluded schizophrenic patients, when compared
to non-deluded non-schizophrenic psychiatric controls and normal controls,

were found to gather less information before making a judgement (that is,
waited for fewer beads to be drawn) and they were found to be overconfident
in rating the correctness of their decisions. Subsequent to that initial study, the
performances of schizophrenic deluded patients, non-schizophrenic deluded
patients (with diagnoses of delusional disorder), anxious controls and normal
controls have been compared on a similar task (Garety, Hemsley and Wesseley,
1991). Both groups of deluded subjects (whether schizophrenic or not) were
found to show the same general pattern of an overconfident jumping-to-con-
clusions style of belief formation when compared to the two control groups.
Of further note, the Garey et al. (1991) study also found that their deluded
patients were more likely than controls to change their hypotheses when
shown disconfirmatory evidence—that is, after having stated their hypothesis
about which jar the beads were being drawn from, deluded patients were more
likely and quicker to switch to saying that it must be the other jar when shown
a new draw of a coloured bead which went against their stated hypothesis.
Similar patterns of performance have also been found in non-clinical delusion-
prone individuals (Linney, Peters and Ayton, 1998).
Garety and Freeman (1999) have recently carried out an extensive review
of probabilistic reasoning and other forms of reasoning in deluded subjects, in
particular patients with diagnoses of schizophrenia and delusional disorder, and
delusion-prone non-clinical individuals. These authors conclude that deluded,
or delusion-prone, individuals show a jumping-to-conclusions style of belief
formation when allowed to decide how much information is needed before
reaching a decision (e.g. number of draws of a bead). In contrast, when subjects
are given a fixed number of trials and asked to make probability estimates,
deluded subjects do not differ, in general, from control subjects. In other
words, Garety and Freeman (1999) suggest that deluded subjects have a data-
gathering bias rather than a probabilistic reasoning deficit. Furthermore, a
number of studies have found evidence of a disconfirmatory bias in deluded
individuals—that is, people with delusions not only appear more ready to jump
to conclusions on the basis of little data, they also appear more ready to aban-
don an existing hypothesis and form a new one on the basis of less first-hand
evidence which runs counter to their current belief. This last set of results is
of some note since our working definition of delusions is that these are false
beliefs which are resistant to counter-evidence. Perhaps what counts for
deluded subjects is the nature of that counter-evidence. We will come back
to that point later.
In sum, delusions are found in the absence of marked intellectual deterio-
ration. Furthermore, deluded individuals show relatively intact logical deduct-
ive reasoning—a finding which is compatible with clinical observations of
schizophrenic patients who have highly systematized delusional systems with
their own idiosyncratic internal logic—and they are perfectly capable of mak-
ing accurate judgements of probability when given all of the relevant infor-
mation. Overall then it seems that Maher is right in saying that a failure of
rational and/or logical thinking is not characteristic of deluded patients. Rather
than an abnormality of reasoning, it seems that what is characteristic of deluded
individuals is an abnormality in the way that they treat hypothesis-generat-
ing evidence.
6. Cognitive Accounts of the Presence of Delusional Beliefs

Based on the empirical evidence reviewed above, Garety and Hemsley (1994)
have proposed that what is critical for explaining the presence of a delusional
belief (rather than the thematic content of a particular delusional belief) is that
deluded individuals have a bias towards focusing on current stimuli when for-
ming beliefs, rather than being influenced by past regularities of experience.
In other words, the notion here is that what sways judgement, or belief, in
the deluded individual is what is happening here and now, rather than what
might have happened in the past, or indeed what might happen in the future.
Stone and Young (1997) have come to a somewhat similar conclusion in their
account of circumscribed delusions, in particular Capgras delusion. More
specifically, these latter authors propose that normal belief formation involves
weighing up hypotheses that are observationally adequate—that is, which equ-
ate with the evidence of our own senses—against hypotheses which fit con-
servatively within our current web of beliefs about how the world should be.
These two need not pull against each other in the normal run of things, but
in the case of Capgras patients they do. And so, it is argued, Capgras delusion
arises when a perceptual aberration (loss of affective responsiveness to a loved
one’s face) is misattributed in the context of an externalizing bias for negative
events—these two factors explain the specific content of the delusion—and
then an additional reasoning bias is needed to explain why these individuals
become deluded. The nature of that reasoning bias is that these individuals
tend to put more weight on forming a belief which satisfies the evidence of
their own senses rather than forming a belief which fits more conservatively
within their current stock of beliefs about the world, but which is not as
observationally adequate.
Thus, both Garety and Hemsley (1994) and Stone and Young (1997) have
proposed bias accounts of delusion formation—that is, the presence of a
delusional belief is explained by the presence of a particular reasoning bias.
For Garety and Hemsley (1994), that bias is conceptualized primarily in terms
of a conflict between different sources of information (current information
versus past and/or future information), whereas Stone and Young (1997)
conceptualize bias more in terms of a conflict between the different ‘purposes’
that beliefs play in our lives. That is, on their view, the conflict is between
beliefs that adequately address the evidence of our own senses versus beliefs
that fit within our current web of beliefs about how the world should be
without the need for major internal reorganization.
Although there is much to be said for bias accounts of delusion formation

(see Stone and Young, 1997), it is our opinion that a deficit model will offer
a more plausible account of delusions, in particular bizarre delusions which
defy commonsense and appear impervious to rational counter-argument.
Bentall (1995) has defined a cognitive deficit as an inability (which might
occur to varying degrees) to perform a certain kind of information-processing
operation, whereas a cognitive bias is a tendency to use information in one
way rather than another way. If the formation of a delusional belief were just
a question of bias, then surely the weight of counter-arguments would eventu-
ally tip the scales in favour of rationality. But, if, instead, normal belief evalu-
ation is impaired in deluded individuals because of an inability to critically
evaluate competing causal explanations in the light of general knowledge and
our broader web of beliefs, then no amount of rational counter-arguments
coming from others, including respected doctors, will dislodge a belief once
it has been adopted. In this case, deluded individuals might find themselves
holding quite absurd beliefs, given their web of general knowledge, and feeling
just as confused as everyone else as to why in all good sense they should believe
such things—as appears to be the case in some deluded patients (Alexander,
Stuss and Benson, 1979).5
If, as Garety and Hemsley (1994) and Stone and Young (1997) argue,
deluded individuals are prone to explanations of one type (explanations that
derive from the immediate evidence of our own senses) over explanations of
another type (explanations that conform to past regularities of experience, gen-
eral knowledge and our broader web of beliefs), then that implies, if we think
in terms of deficit rather than bias, that normal belief evaluation may depend
upon the suspension of automatically prioritized explanatory biases, which
reflect, in part, the relative salience of different sources of information:
(1) first-person information which we access directly via our senses;

(2) second-person information which others tell us about or that we infer
from others’ actions;
(3) third-person information about the world that we have acquired
through learning and past experience.
On this view, normal rational evaluation of the plausibility and probability of

competing explanations is sustained by a more fundamental ability to ‘re-
initialize’ explanations that derive from different sources of information in
order to give equal priority to hypotheses based on first-person, second-person
5
Note here that we are not suggesting that all delusions are fixed and stable. Delusions can
come and go and may respond to anti-psychotics. Rather, what we are suggesting is that
the presence of a delusional belief reflects a failure of normal belief evaluation which is more
fundamentally caused by some neuropathy and that neuropathy may or may not be part of
an unstable ‘disease process’ which is responsive to medication.
and third-person evidence. Given this, then we would also need to argue that
it is a normal human bias to favour explanations that derive from first-person
sensory information over explanations that derive from indirect second-person
and third-person information. But that sounds quite reasonable. It would be
surprising if we did not default to first believing the evidence of our own
senses. After all, there would be little evolutionary advantage in doubting the
evidence of our own senses—much safer to act first on the basis of directly
perceived danger and then evaluate whether or not we were misguided if
prompted to do so later by conflicting second-person and/or third-person
information. Furthermore, subsumed within this natural human bias to favour
hypotheses based on first-person evidence would be that other universal bias,
referred to earlier, to favour personal-level causal explanations over subper-
sonal-level causal explanations. Since we do not have direct sensory access to
subpersonal states, subpersonal causal explanations will always depend upon
second-person and/or third-person information and hence will always be low
on our priority listing of possible explanations, if they make the list at all. And
finally, individual biases of attributional style would also serve to prioritize
certain types of causal explanations for a particular individual.
In sum, Maher is right, we think, in arguing that it is not irrational to
generate a bizarre causal hypothesis, relative to what others believe, when con-
fronted by a bizarre perceptual experience which does not accord with the
way that the rest of society experiences reality. The nature of the perceptual
aberration, nuanced by an individual’s attributional bias, explains the content of
the favoured causal hypothesis generated to explain what is happening. How-
ever, something else is needed to explain why that implausible hypothesis is
then uncritically adopted and maintained as belief. We think that bias accounts
which attempt to explain the adoption of delusional beliefs in terms of the
presence of a particular type of individual reasoning bias are inadequate to
explain the presence of bizarre delusions which persist despite overwhelming
rational counter-argument. Instead, we advocate a deficit model. That
is, implausible hypotheses are uncritically adopted as delusional beliefs when
there is damage to the normal cognitive system of belief evaluation. In parti-
cular, we suggest that normal belief evaluation depends fundamentally upon
an ability to suspend automatically prioritized explanatory biases, which reflect,
in part, the relative salience of different sources of information: first-person
information which we access directly via our senses; second-person infor-
mation which others tell us about or that we infer from others’ actions; and
third-person information about the world acquired through learning and past
experience. To put it another way in order to clarify the bias versus deficit
distinction, prioritizing causal explanations that are based on direct first-person
evidence over explanations that are based on indirect sources of information—
the latter of which necessarily include subpersonal-level causal explanations—
is something which all people do; it is not something which only certain people
with a particular type of reasoning bias do. Furthermore, there is something
else which all people can do, given that everything is operating normally, and
that is to access some override safety mechanism which has the job of sus-
pending automatic biases in order to critically evaluate different hypotheses,
‘re-initialized’ as having equal priority whether based on direct or indirect
sources of information. Damage to this normal safety-check mechanism (a
deficit) is necessary, we think, to explain the presence of delusional beliefs.
In the following section we outline, in more detail, a model of the normal
cognitive system of belief generation and belief evaluation which we believe
is applicable generally to the explanation of all delusions as being caused by
damage to one or more components of that normal system.
7. A Cognitive Model of Normal Belief Generation and

Evaluation
We propose that delusion formation is best explained in relation to a three-
factor model of normal belief generation and belief evaluation which incorpor-
ates the following features: (1) there is current information about self and the
environment, provided by sensory mechanisms, which requires explanation;
(2) universal and individual attributional biases exist which influence the gener-
ation of favoured causal explanations; and (3) there are normal belief evaluation
processes which suspend the natural favoured status of first-person evidence
in order to critically evaluate the plausibility and probability of all hypotheses,
given equal priority whether they derive from direct first-person experience
or indirect sources of information. The following is a very tentative proposal
of how these factors might interact in the normal system of belief formation.6
Belief formation begins with current sensory information that is being pro-
vided about self and the environment. Some pieces of current sensory infor-
mation will be promoted as requiring explanation. These seem to be of two
types: (1) there is information which we selectively attend to because of height-
ened personal salience; and (2) there is information which we are automatically
oriented towards because it is discordant with our prior experience of how
the world should be, whether that discordance be due to perceptual aberrations
or due to the occurrence of novel environmental stimuli. It may be that two
distinct mechanisms carry out these two respective monitoring functions: (1)
a conscious monitoring operation which matches present sensory inputs against
sets and expectancies that derive from beliefs of which we can be conscious;
and (2) a preconscious monitoring operation which detects mismatches against
past regularities of experience. The proposition that two such monitoring
mechanisms orient our attention to different sensory information is consistent
with Ben-shakhar’s (1994) proposal that autonomic responsiveness to signifi-
6
Note that the early stages of this framework are very similar to Garety and Hemsley’s (1994)
model of delusion formation.
cant (personally relevant) stimuli is controlled by a mechanism which differs

from the mechanism that directs autonomic orientation to novel nonsignificant
changes. The purpose of monitoring is to identify discordant sensory data,
novel sensory data and sensory data with heightened personal salience which
then requires explanation.
Given the nature of what needs to be explained, hypotheses based on rel-
evant data from different sources are generated: other first-person sensory
information; second-person information derived from others; and third-person
information derived from our broader web of beliefs, general knowledge
(stored in semantic memory) and information from prior experience (stored in
episodic memory). Note here the contrast between other first-person sensory
information which is available to sufferers of phantom limb experiences and
that which is directly available to Capgras patients. Capgras patients only have
the single aberrant perception—‘he looks right but he doesn’t feel right’: there
is no other directly accessible first-person information available. In contrast,
individuals with the phantom-limb experience—‘my arm feels like it’s there
but I can’t see it’—can try and fail to move or touch their arms in order to
collect other relevant first-person information.
Based on these various sources of evidence, a prioritized list of possible
causal hypotheses is generated. There are at least two different biases which
play a role in the generation of an individual’s prioritized list of explanations.
First, there are individual differences in attributional bias which will influence
the particular type of causal hypothesis that an individual will favour. And
second there is a universal attributional bias which we all have and that is to
favour personal-level causal explanations that satisfy the evidence of our own
senses over explanations that are based on second-person and third-person
information, which would include subpersonal-level causal explanations.
These prioritized candidates for belief are then submitted to rational evalu-
ation. Rational belief evaluation processes are underpinned by an ability to
suspend personalized and individualized priority weightings of generated
hypotheses. Re-initializing candidates for belief in this way may involve
source-referencing explanations in order to give equal priority to ‘what I think
is happening’, ‘what others say is happening’ and ‘what the doctor says has
gone wrong in my brain to cause me to have this experience’. Once hypotheses
have been re-initialized as having equal priority, critical evaluation and rationa-
lization processes can proceed. These processes may include assessing how con-
sistent different hypotheses are with our broader web of beliefs, as suggested
by Stone and Young (1997), and perhaps some probabilistic assessment of how
likely particular hypotheses are—along the lines of ‘Has this happened to me
before, and if so what caused it then?’ or ‘When this happens to others what
do they think is wrong?’
After the critical evaluation process, a particular hypothesis may be adopted
as belief and that belief is then added to our current web of beliefs along with
any necessary revision of the other beliefs which we hold. The updated web
of beliefs in turn influences how we act and what new expectations we have
about incoming sensory information. Alternatively, the result of the rational
evaluation process may be to defer selection of any particular hypothesis, on
the basis that some other, as yet unspecified, hypothesis might be generated
in the future if I were to take the time to collect more information. And note
here that difficulty giving equal priority to the possibility that such a future
alternative hypothesis might be generated on the basis of as-yet unknown data
could lead to less data-gathering and a tendency to jump to conclusions on
the basis of immediate first-person evidence. This is entirely consistent with
the empirical evidence of a jumping-to-conclusions style of belief formation
and a disconfirmatory bias in deluded individuals.
In the case of deluded individuals, normal belief evaluation processes cannot
be completed because these individuals have difficulty suspending natural biases
in order to give equal priority to all possible hypotheses, regardless of their
primary source of information. Consequently, deluded individuals will adopt
as belief whichever hypothesis is their natural highest priority explanation—
and that will always be a personal-level causal explanation based on first-person
experience, and it will always reflect an individual’s particular causal attri-
butional bias.
But it is important to bear in mind here that the naturally favoured causal
explanation, generated to explain novel and/or relevant sensory information,
will not necessarily be incorrect, even in individuals with a failure of normal
belief evaluation. If there is nothing wrong with the neural substrates which
provide us with sensory information about the environment (including our
own body states) and/or orient our attention towards relevant aspects of the
environment, then there is no reason to think that our natural biases will lead
us markedly astray. However, they will most certainly lead us astray if a percep-
tual aberration is present. Hence, we argue that in order to explain bizarre
delusions which defy everyday commonsense two components of the normal
three-factor model must be damaged: (1) there must be some disruption to
the normal operation of sensory and/or attentional mechanisms which provide
us with sensory information and/or orient our attention to relevant stimuli
within the environment; and (2) there must also be a failure of normal
belief evaluation.
In the following section we illustrate how a two-deficit model of delusion
formation, incorporating both a perceptual aberration (defining perception
very broadly to encompass phenomenological experiences which derive from
primary sensory information—sight, hearing, touch, smell, taste—as well as
somatic information about body states, and perhaps even emotional states) and
a deficit of normal belief evaluation, when mediated by (normal) individual
attributional biases, can explain a diversity of bizarre delusions including cases
of ‘organic’ delusions and cases of ‘psychotic’ delusions.

8. Bizarre Two-Deficit Delusions
8.1 Capgras and Cotard Delusion

Both Capgras delusion and Cotard delusion arise when normal belief evalu-
ation is impaired in individuals who are experiencing the aberrant percep-
tion—‘he looks right but he doesn’t feel right’—which, as argued earlier, arises
when an individual loses the sense of affective responsiveness which normally
co-occurs with explicit recognition of a familiar face. In the case of Capgras
patients, who have an attributional bias to externalize the cause for negative
events, the naturally favoured personal-level causal explanation which is then
generated and adopted as belief is the bizarre delusion that an impostor has
replaced their loved one. In contrast, Cotard patients with an attributional bias
to internalize the cause for negative events will generate and adopt as belief
the naturally favoured personal-level causal explanation that they are dead
(Wright et al., 1993; Young et al., 1993).
8.2 Alien Control Delusions

Schizophrenic delusions of alien control arise when normal belief evaluation
fails in individuals who experience a discordant sense of passivity which arises
when normal on-line self-monitoring is disrupted, resulting in loss of the sense
of self-generation that normally accompanies self-initiated action. Although no
studies have investigated the causal attributional styles of patients with delusions
of alien control, it seems most likely that this delusion is the naturally favoured
personal-level causal explanation of individuals who have an attributional bias
to externalize the cause for negative events. In contrast, deluded individuals
who experience this perceptual aberration and who have an attributional bias
to internalize the cause for negative events might be expected to develop nihil-
istic delusions of soul-death, for example.
8.3 Auditory Hallucinations and Loss of Boundary Delusions

Frith (1992) has proposed that normal internal monitoring of self-generation
is not only required for a sense of ownership of bodily actions (as described
above) but is also required for a sense of ownership of inner thoughts. If self-
generation in this second sense is lost, then Frith (1992) proposes that affected
individuals will experience their inner thoughts with a sense of unintendedness
and alien influence. That is, inner thoughts will be experienced as extrapersonal
auditory perceptions—as if one is ‘hearing voices’. But, in line with what we
have argued before, it is possible to have the experience of ‘hearing voices’
and yet know that for some reason—not directly accessible—the source of
those voices is oneself and not some external agency. The experience of ‘hear-
ing voices’ is not specific to patients with a diagnosis of mental illness. Auditory
hallucinations are experienced by individuals with ear disease and can be
induced in normal subjects using sleep deprivation, isolation/sensory depri-
vation and, in some cases, simple suggestion (see Bentall, 1990, and David,
1994, for reviews). And so we think that auditory hallucinations, when these
are reported as the voices of real entities (as in patients with schizophrenia),
are just another sub-category of delusion—an aberrant perceptual experience
being delusionally misinterpreted in the absence of normal belief evaluation.
Individual differences in attributional bias might then manifest in the differ-
ent motives which patients ascribe to the sources of their voices. For example,
some patients, perhaps those with a bias towards internalizing blame for nega-
tive events, may believe that the voices are there to punish them for wrongs
that they have done; whilst others, perhaps those with a bias towards exter-
nalizing blame for negative events, might believe themselves to be innocents
being ‘picked upon’ by malevolent forces.
But then again, attributional biases may play a more fundamental role here.
That is, if an individual has lost the sense of self-generation which normally
tags inner thoughts as self-sourced, then, if that individual has an externalizing
bias for negative events, perhaps it is only thoughts with unpleasant (negative)
content which are misattributed to external voices. Voices need not have nega-
tive content—there are many non-clinical individuals who hear helpful voices
which they interpret positively as spirit guides (Romme and Escher, 1989).
However, in the case of schizophrenia, many patients who hear voices are also
paranoid and, in such cases, voices are usually experienced as unpleasant, criti-
cal and even threatening. Since paranoia is known to be associated with an
externalizing bias for negative events, what may be happening in cases of para-
noid schizophrenic patients who hear ‘negative’ voices is that an externalizing
bias for negative events interacts with loss of the normal ability to tag self-
generated thoughts as one’s own, thus causing the misattribution to external
voices of thoughts with ‘negative’, rather than ‘positive’, content.
Up until now, we have only talked about one type of perceptual aberration
being implicated in the presence of auditory hallucinations—that is, loss of the
sense of self-generation associated with one’s thoughts. However, whereas
Frith and colleagues (Frith, 1992; Frith and Done, 1989) interpret auditory
hallucinations and other ‘loss of boundary’ experiences—thought-broadcasting,
mind-reading, and thought-insertion—as all being straightforward manifes-
tations of a source-monitoring failure, David (1994) points out that this may
only be part of the story. That is, David suggests that two perceptual aber-
rations, in combination, may be needed to explain the phenomenology of
auditory hallucinations—a source-monitoring defect coupled with the loss of
subjective privacy normally associated with ‘hearing in mind’ one’s inner
thoughts. Recall that earlier we spoke about how loss of a sense of subjective
privacy when coupled with different styles of attributional bias might explain
why some patients generate different types of ‘loss of boundary’ delusions.
However, David suggests that different ‘loss of boundary’ experiences might
be better understood as reflecting different types of perceptual aberration, in
conjunction or in isolation. That is, whereas auditory hallucinations may reflect
a source-monitoring defect coupled with the loss of subjective privacy nor-

mally associated with ‘hearing in mind’ one’s thoughts, thought-insertion
delusions may reflect a source-monitoring defect coupled with an intact sense
of ‘hearing thoughts’ within subjective space. In other words, patients with
thought-insertion delusions may ‘hear’ their thoughts intrapersonally but not
as self-instigated. In contrast, there may be other patients who have lost the
sense of subjective privacy associated with ‘hearing in mind’ one’s thoughts,
and so experience their thoughts extrapersonally, but with an intact sense of
self-generation. These patients may then generate delusions of thought-broad-
casting.
8.4 Delusions of Reference

Delusions of reference occur when a deluded individual reads personal signifi-
cance into seemingly trivial comments and activities of others and into com-
pletely unrelated and commonplace events. For example, patients with
delusions of reference may believe that statements being made by television
or radio commentators are specifically directed at themselves or they may
believe that colours and arrangements of everyday objects have been specifi-
cally arranged in such a way so as to convey hidden messages to them.
Although there is no empirical basis for assuming that a perceptual aberration
plays a role in the generation of delusions of reference, the phenomenological
self-reports of patients do attest to some perceptual quality being present. For
example, one patient described the experience as a feeling that things and
people ‘zoomed in’ on them with an overwhelming sense of self-referential
import.
Let’s assume, for the sake of argument, that there is some perceptual basis
to delusions of reference and let us now consider how that aberration might
interact with different attributional styles. Delusions of reference often co-
occur with paranoid delusions in patients with schizophrenia. In these cases,
it seems likely that the aberrant perception is interpreted negatively and then
coupled with an attributional bias to externalize blame for negative events.
The result, given a deficit of normal belief evaluation, is that these individuals
will generate and adopt as belief some form of conspiracy theory about being
followed and electronically monitored by agents of forces who are plotting
against them. In contrast, delusions of reference can also be found to co-occur
with grandiose delusions and patients with grandiose delusions are known to
have the same attributional bias as paranoid patients (that is, a tendency to
externalize blame for negative events and internalize blame for positive events).
In these cases then, it seems likely that the same perceptual aberration (to that
experienced by paranoid patients with delusions of reference) is interpreted
positively and then attributed to self, thus generating grandiose delusions about
why that person should be in the spotlight or see significance in events which
others deem commonplace. For example, these individuals might develop
delusions about having psychic powers or about being chosen specially by God
to receive secret communications.
8.5 Hallucinations
Up until now, we have focused on auditory hallucinations, specifically of the
voices commenting type, which are said to arise when normal monitoring of
internal speech fails (Frith, 1992). However, hallucinations can occur in any
modality: auditory, visual, tactile, olfactory, and gustatory. Furthermore, hal-
lucinations in various modalities can co-occur in the same individuals. In such
cases then, it seems more plausible that these hallucinations reflect difficulty
discriminating self-generated imaginary states from externally generated sensory
experiences (Bentall, 1990).
Now consider an individual who has this type of perceptual aberration
(multi-modal hallucinations caused by defective discrimination of imagination
and reality) and a deficit of normal belief evaluation. Such an individual’s entire
web of beliefs will quickly get wildly out of kilter: an initial delusion will
prompt imaginings which are then experienced as confirmatory first-person
sensory information for the deluded belief and then further elaborative imagin-
ings will lead to other delusions, and so forth and so forth. It is therefore
plausible that one cause of polythematic, or fragmentary, delusions found to
occur in cases of schizophrenia is the coupling of a deficit of normal belief
evaluation with a single perceptual aberration, which by its very nature (a
failure to discriminate self-generated imaginary states from externally generated
sensory experiences) has a cascading effect on the individual’s entire web of
beliefs. However, even here we would argue that if a person is capable of
normal belief evaluation, then they will be able to check this disintegration
process by taking on board what others are telling them and, thus prompted,
test out the plausibility of their wild imaginings in the light of learned gen-
eral knowledge.
8.6 Non-Hallucinatory Polythematic Delusions

Earlier in the paper, we noted that any full account of delusions must be able
to account for both the diversity of content and the diversity of scope in
delusions. In the previous section, we suggested that one cause of widespread
polythematic delusions might be the special case of a deficit of normal belief
evaluation coupled with multi-modal hallucinations caused by a failure to dis-
criminate reality from imagination. Now we consider other possible causes of
non-hallucinatory polythematic delusions.
A possible explanation, which is consistent with what has gone before, is
that polythematic delusions may arise when a failure of normal belief evaluation
is coupled with a multitude of different perceptual aberrations. This might
occur in two ways. First, we have already seen that specific schizophrenic
delusions may arise, in part, in response to specific perceptual aberrations—
for example, loss of boundary experiences and delusions of reference. Now,

given the range of possible perceptual aberrations and the extent of cognitive
deficits which have been found in patients with schizophrenia, including, for
example, face recognition deficits (Ellis and Young, 1996) and memory deficits
(McKenna et al., 1990), it is likely that some cases of polythematic delusions
occur when patients are faced with a multitude of different perceptual aber-
rations which are all jockeying for explanation. However, there is a second
possibility. We have already hinted at the idea that anomalous perceptual
experiences might arise either because there is dysfunction of one or more
sensory mechanisms dedicated to providing certain types of sensory infor-
mation (e.g. face-recognition information) or because the normal attentional
mechanisms which orient us to novel and/or relevant stimuli in the environ-
ment are themselves dysfunctional (see Maher and Spitzer, 1993, for a review
of abnormal attention and orienting in patients with schizophrenia). If this
latter type of perceptual aberration is present, then a multitude of different
events which would normally be ignored as mere commonplace might be
imbued with an aberrant quality of salience. That idea accords well with
schizophrenic patients’ self-reports of a heightened sense of significance and
meaningfulness accompanying a myriad of everyday events that might other-
wise be deemed commonplace or coincidental (Anscombe, 1987). In other
words, the idea here is that instead of experiencing a set of perceptual aber-
rations which each reflect a distinct sensory dysfunction, some schizophrenic
individuals may be experiencing the world, in general, aberrantly because of
central dysfunction to early attentional/orienting mechanisms which direct us
to events in the world requiring explanation.
The above examples have served to illustrate how two deficits in a three-
component model of normal belief generation and belief evaluation (in which
attributional biases are seen as part of the normal) can explain the diversity of
content and the diversity of scope in various ‘organic’ and ‘psychotic’ bizarre
delusions. Previously, we argued that a single deficit which causes a perceptual
aberration, in the absence of a failure of normal belief evaluation, is not suf-
ficient to cause a delusional belief. In these cases, the non-deluded individual
is capable of adopting ‘as if’ accounts of the type espoused by patients with
depersonalization disorder: ‘it feels as if an alien is controlling my body, but
aliens don’t really exist, and the doctors tell me that I have a chronic abnormal
reaction to stress—so that must be what it is’. In the following section, we
consider what happens if there is just a single deficit of normal belief
evaluation.
9. The Implications of a Single Deficit of Normal Belief

Evaluation
We have already suggested that an individual with a deficit of normal belief
evaluation will not necessarily generate naturally favoured personal-level causal
explanations which are incorrect. However, even our normal sensory and/or
attentional mechanisms are unlikely to operate perfectly all of the time. And
so, there may well be times when we think that we see a person’s shadow or
when events occur with a sense of meaningful contiguity (e.g. someone calls
you on the telephone just as you are thinking about that person). In the case
of a person who has a deficit of normal belief evaluation, and thus difficulty
evaluating the plausibility and probability of generated hypotheses in the light
of past experience and general knowledge, such first-person experiences may
generate beliefs such as thinking that you have seen a ghost or thinking that
you have psychically foreseen the future. But since there is no fixed sensory
and/or attentional dysfunction causing continual perceptual aberrations, then
these will be relatively isolated events. We therefore think it unlikely that such
individuals will generate full-blown delusions which disrupt normal func-
tioning, thus leading to some form of clinical diagnosis. Rather, we think that
these individuals will be more likely to present as idiosyncratic non-clinical
individuals who are prone to hold unorthodox beliefs.
However, we do acknowledge that there may be some instances when a
single deficit of normal belief evaluation can result in delusional beliefs—of an
ordinary type. That is, thus far we have focused mainly on bizarre delusions
which run counter to the commonsense beliefs of others. However, as men-
tioned earlier, there is a class of clinically diagnosable delusions which are not
bizarre. These ‘ordinary’ delusions include, for example, delusions of jealousy,
the hypochondriacal delusion that you are going to die, or the belief that
someone important secretly loves you. Such delusions are diagnosed, usually
as cases of delusional disorder, when an individual exhibits delusional think-
ing—that is, they maintain their beliefs despite the lack of compelling objective
evidence—but also when the delusion is about everyday things. That is, in
these cases, the content of the delusion is not outside the bounds of common-
sense—such things can and do occur. Here we think that what may be hap-
pening is that there is a deficit of normal belief evaluation coupled with misper-
ceptions of partial sensory information and misinterpretations of ambiguous
cues which are driven by heightened, but not abnormal, attentional and
interpretative biases. There is plenty of evidence to suggest that both deluded
patients (Bentall and Kaney, 1989; Bentall, Kaney and Bowen-Jones, 1995;
Kaney et al., 1992; Leafhead, Young and Szulecka, 1996) and non-deluded
individuals with social phobia and anxiety disorders (Matthews and Mackin-
tosh, 1998; McNally, 1998) can be perceptually sensitized to identify certain
stimuli and to preferentially elaborate information which conforms to and
reinforces their particular expectations and preoccupations. In the case of indi-
viduals with a deficit of normal belief evaluation, such misperceptions and
misinterpretations of ambiguous, but certainly not bizarre, first-person evi-
dence may lead to the formation of ordinary delusions.

10. Summary and Conclusions

After reviewing the role that perceptual aberrations and attributional biases
play in determining the thematic content of different delusions, and reviewing
research investigating the presence of abnormal thinking and/or reasoning in
deluded individuals, we propose that the formation of delusional beliefs is best
explained in terms of damage to one or more components of a three-compo-
nent system of normal belief generation and belief evaluation. This three-
component system includes: (1) attentional/monitoring mechanisms which
orient us to current sensory information requiring explanation; (2) universal
and individual attributional biases which influence the generation of naturally
favoured causal explanations; and (3) belief evaluation processes which are
underpinned by an ability to suspend the automatic prioritizing of hypotheses
based on immediate first-person experience in order to critically evaluate
hypotheses, given equal priority whether based on direct or indirect sources
of information.
We have used this model to present a deficit model of delusion formation
rather than a bias model. Bias accounts we think inadequate to explain bizarre
delusions which defy commonsense and persist despite overwhelming rational
counter-argument. More specifically, we propose that damage to the third
component of the normal model results in deluded individuals adopting as
belief whichever hypothesis is naturally favoured as their highest priority per-
sonal-level causal explanation, nuanced by individual attributional biases. This,
we think, helps to explain why some deluded individuals can express their
bizarre belief with a sense of self-evident incontrovertible certainty, whilst at
the same time appearing to be aware on some level (e.g. being confused about
why they should think such an odd thing) of the implausibility of that belief.
Bizarre delusions, we argue, require at least two deficits: (1) at least one
form of perceptual aberration, whether caused by dysfunction of a sensory
mechanism or caused by dysfunction of attentional/orienting mechanisms; and
(2) a breakdown of normal belief evaluation. Thus, the existence of deficits
which cause one or more perceptual aberrations is not sufficient to explain
the formation of delusional beliefs. Furthermore, since naturally favoured causal
explanations generated in the absence of a stable perceptual aberration will,
most often, be correct, a single deficit of normal belief evaluation will not
necessarily lead to full-blown delusions. Such individuals we think more likely
to present as non-clinical idiosyncratic individuals who are prone to unortho-
dox beliefs. However, in cases where an attentional bias, driven by personal
preoccupations and expectations, leads to continual misperceptions of partial
information and misinterpretations of ambiguous stimuli, then a single deficit
of normal belief evaluation may manifest in delusions with ordinary content.
Macquarie Centre for Cognitive Science

Macquarie University

References
Abramson, L.Y., Seligman, M.E.P. and Teasdale, J.D. 1978: Learned helplessness in
humans: Critique and reformulation. Journal of Abnormal Psychology, 87, 49–74.
Alexander, M.P., Stuss, D.T. and Benson, D.F. 1979: Capgras syndrome: A reduplicat-
ive phenomenon. Neurology, 29, 334–9.
American Psychiatric Association 1994: Diagnostic and Statistical Manual of Mental Dis-
orders, Fourth Edition (DSM-IV). Washington, DC: American Psychiatric Associ-
ation.
Angel, R.W. 1976: Efference copy in the control of movement. Neurology, 26, 1164–8.
Anscombe, R. 1987: The disorder of consciousness in schizophrenia. Schizophrenia
Bulletin, 13, 241–60.
Ben-shakhar, G. 1994: The roles of stimulus novelty and significance in determining
the electrodermal orienting response: Interactive versus additive approaches. Psych-
ophysiology, 31, 402–11.
Benson, D.F., Gardiner, H. and Meadows, J.C. 1976: Reduplicative paramnesia. Neur-
ology, 26, 147–51.
Bentall, R.P. 1990: The illusion of reality: A review and integration of psychological
research on hallucinations. Psychological Bulletin, 107, 82–95.
Bentall, R.P. 1995: Brains, biases, deficits and disorders. British Journal of Psychiatry,
167, 153–5.
Bentall, R.P. and Kaney, S. 1989: Content-specific processing and persecutory
delusions: an investigation using the emotional Stroop test. British Journal of Medical
Psychology, 62, 355–64.
Bentall, R.P. and Kaney, S. 1996: Abnormalities of self-representation and persecutory
delusions: A test of a cognitive model of paranoia. Psychological Medicine, 26,
1231–7.
Bentall, R.P., Kaney, S. and Bowen-Jones, K. 1995: Persecutory delusions and recall
of threat-related, depression-related, and neutral words. Cognitive Therapy and
Research, 19, 445–57.
Bentall, R.P. Kaney, S. and Dewey, M.E. 1991: Paranoia and social reasoning: An
attribution theory analysis. British Journal of Clinical Psychology, 31, 12–23.
Bentall, R.P. Kinderman, P. and Kaney S. 1994: The self, attributional processes and
abnormal beliefs: Towards a model of persecutory delusions. Behaviour Research
and Therapy, 32, 331–41.
Candido, C.L. and Romney, D.M. 1990: Attributional style in paranoid vs depressed
patients. British Journal of Medical Psychology, 63, 355–63.
Coltheart, M. 1984: Editorial. Cognitive Neuropsychology, 1, 1–8.
Cutting, J. 1985: The Psychology of Schizophrenia. Edinburgh: Churchill Livingston.
Damasio, A.R., Tranel, D. and Damasio, H. 1991: Somatic markers and the guidance
of behaviour: Theory and preliminary testing. In H.S. Levin, H.M. Eisenberg and
A.L. Benson (eds), Frontal Lobe Function and Dysfunction. Oxford University Press,
217–29.
David, A.S. 1994: The neuropsychological origin of auditory hallucinations. In A.S.
David and J.C. Cutting (eds), The Neuropsychology of Schizophrenia: Brain Damage,
Behaviour and Cognition Series. Hove, E. Sussex: Lawrence Erlbaum Associates,
269–313.
Davis, K.D., Kiss, Z.H., Luo, L., Tasker, R.R., Lozano, A.M. and Dostrovsky, J.O.
1998: Phantom sensations generated by thalamic microstimulation. Nature, 391,
385–7.
Davison, G.C. and Neale, J.M. 1998: Abnormal Psychology 7th edn. New York: Wiley.
de Pauw, K.W., Szulecka, T.K. and Poltock, T.L. 1987: Fregoli syndrome after cer-
ebral infarction. Journal of Nervous and Mental Disease, 175, 433–8.
Elliott, R. and Sahakian, B.J. 1995: The neuropsychology of schizophrenia: Relations
with clinical and neurobiological dimensions. Psychological Medicine, 25, 581–94.
Ellis, A.W. and Young, A.W. 1988: Human Cognitive Neuropsychology. Hove, E. Sussex:
Lawrence Erlbaum Associates.
Ellis, H.D. and Young, A.W. 1990: Accounting for delusional misidentifications. Bri-
tish Journal of Psychiatry, 157, 239–48.
Ellis, H.D. and Young, A.W. 1996: Problems of face perception in schizophrenia. In
C. Pantelis, H.E. Nelson and T.R.E. Barnes (eds), Schizophrenia: A Neuropsycho-
logical Perspective. New York: Wiley, 397–416.
Ellis, H.D., Young, A.W., Quayle, A.H. and de Pauw, K.W. 1997a: Reduced auto-
nomic responses to faces in Capgras delusion. Proceeding of the Royal Society of
London: Biological Sciences, B264, 1085–92.
Ellis, H.D., Young, A.W., Quayle, A.H. and de Pauw, K.W. 1997b: Response from
Ellis, Young, Quayle and de Pauw. Trends in Cognitive Sciences, 1, 158.
Filley. C.M. and Jarvis, P.E. 1987: Delayed reduplicative paramnesia. Neurology, 37,
701–3.
Fleminger, S. and Burns, A. 1993: The delusional misidentification syndromes in
patients with and without evidence of organic cerebral disorder: A structured
review of case reports. Biological Psychiatry, 33, 22–32.
Flor, H., Elbert, T., Muhlnickel, W., Pantev, C., Wienbruch, C. and Taub, E. 1998:
Cortical reorganization and phantom phenomena in congenital and traumatic
upper-extremity amputees. Experimental Brain Research, 119, 205–12.
Freud, S. 1963: Three Case Histories. New York: Collier Books/Macmillan Pub-
lishing Co.
Frith, C. 1992: The Cognitive Neuropsychology of Schizophrenia. Hove, E. Sussex: Lawr-
ence Erlbaum Associates.
Frith, C. and Done, D.J. 1986: Routes to action in reaction time tasks. Psychological
Research, 48, 169–77.
Frith, C. and Done, D.J. 1989: Experiences of alien control in schizophrenia reflect
a disorder in the central monitoring of action. Psychological Medicine, 19, 359–63.
Garety, P.A. and Freeman D. 1999: Cognitive approaches to delusions: A critical
review of theories and evidence. British Journal of Clinical Psychology, 38, 113–54.
Garety, P.A. and Hemsley, D.R. 1994: Delusions: Investigations into the Psychology of
Delusional Reasoning. Oxford University Press.
Garety, P., Hemsley, D.R. and Wesseley, S. 1991: Reasoning in deluded and schizo-
phrenic and paranoid patients. Journal of Nervous and Mental Disease, 179, 194–201.
Graham, S. and Folkes, V.S. 1990: Attribution theory: Applications to achievement,

mental health, and interpersonal conflict. Hillsdale, NJ: Lawrence Erlbaum Associ-
ates.
Hakim, H., Verma, N.P. and Greiffenstein, M.F. 1988: Pathogenesis of reduplicative
paramnesia. Journal of Neurology, Neurosurgery, and Psychiatry, 51, 839–41.
Halligan, P.W., Marshall, J.C. and Wade, D.T. 1995: Unilateral somatoparaphrenia
after right hemisphere stroke: A case description. Cortex, 31, 173–82.
Heider, F. 1944: Social perception and phenomenal causality. Psychological Review, 51,
358–74.
Heider, F. 1958: The Psychology of Interpersonal Relations. New York: Wiley.
Hewstone, M. 1983: Attribution Theory: Social and Functional Extensions. Oxford:
Blackwell.
Hewstone, M. 1989: Causal attribution: From Cognitive Processes to Collective Beliefs.
Oxford: Blackwell.
Huq, S.F., Garety, P. and Hemsley, D.R. 1988: Probabilistic judgements in deluded
and non-deluded subjects. Quarterly Journal of Experimental Psychology, 40A, 801–
12.
Kaney, S. and Bentall, R.P. 1989: Persecutory delusions and attributional style. British
Journal of Medical Psychology, 62, 191–8.
Kaney, S., Wolfenden, M., Dewey, M.E. and Bentall, R.P. 1992: Persecutory
delusions and the recall of threatening and non-threatening propositions. British
Journal of Clinical Psychology, 31, 85–7.
Kasinin, J.S. 1944: Language and Thought in Schizophrenia. Berkeley: University of Cali-
fornia Press.
Kemp, R., Chua, S., McKenna, P. and David, A. 1997: Reasoning and delusions.
British Journal of Psychiatry, 170, 398–405.
Kinderman, P. and Bentall, R.P. 1996a: A new measure of causal locus: The internal,
personal, and situational attributions questionnaire. Personality and Individual Differ-
ences, 20, 261–4.
Kinderman, P. and Bentall, R.P. 1996b: Self-discrepancies and persecutory delusions:
Evidence for a model of paranoid ideation. Journal of Abnormal Psychology, 105,
106–13.
Kinderman, P. and Bentall, R.P. 1997: Causal attributions in paranoia and depression:
Internal, personal, and situational attributions for negative events. Journal of Abnor-
mal Psychology, 106, 341–5.
Kinderman, P., Kaney, S., Morley, S. and Bentall, R.P. 1992: Paranoia and the defens-
ive attributional style: Deluded and depressed patients’ attributions about their
own attributions. British Journal of Medical Psychology, 65, 371–83.
Leafhead, K.M., Young, A.W. and Szulecka, T.K. 1996: Delusions demand attention.
Cognitive Neuropsychiatry, 1, 5–16.
Leudar, I., Thomas, P. and Johnston, M. 1994: Self-monitoring in speech production:
Effects of verbal hallucinations and negative symptoms. Psychological Medicine, 24,
749–61.
Linney, Y.M., Peters, E.R. and Ayton, P. 1998: Reasoning biases in delusion-prone
individuals. British Journal of Clinical Psychology, 37, 285–302.
Lyon, H.M., Kaney, S. and Bentall, R.P. 1994: The defensive function of persecutory
delusions: evidence from attribution tasks. British Journal of Psychiatry, 164, 637–46.
Maher, B.A. 1974: Delusional thinking and perceptual disorder. Journal of Individual
Psychology, 30, 98–113.
Maher, B.A. 1988: Anomalous experience and delusional thinking: The logic of expla-
nations. In T.F. Oltmanns and B.A. Maher (eds), Delusional Beliefs. New York:
Wiley, 15–33.
Maher, B.A. 1992: Delusions: Contemporary etiological hypotheses. Psychiatric Annals,
22, 260–68.
Maher, B.A. and Ross, J.S. 1984: Delusions. In H.E. Adams and P.B. Sutker (eds),
Comprehensive Handbook of Ppsychopathology. New York: Plenum Press.
Maher, B.A. and Spitzer, M. 1993: Delusions. In C.G. Costello (ed.) Symptoms of
Schizophrenia. New York: Wiley.
Matthews, A. and Mackintosh, B. 1998: A cognitive model of selective processing in
anxiety. Cognitive Therapy and Research, 22, 539–60.
McKenna, P.J., Tamlyn, D., Lund, C.E., Mortimer, A.M., Hammond, S. and Badde-
ley, A.D. 1990: Amnesic syndrome in schizophrenia. Psychological Medicine, 20,
967–72.
McNally, R.J. 1998: Experimental approaches to cognitive abnormality in posttrau-
matic stress. Clinical Psychology Review, 18, 971–82.
Peterson, C., Semmel, A., von Bueyer, C., Abramson, L.Y., Metalsky, G.I. and Selig-
man, M.E.P. 1982: The attributional style questionnaire. Cognitive Therapy and
Research, 6, 287–300.
Ramachandran, V.S. and Hirstein, W. 1998: The perception of phantom limbs. The
D.O. Hebb lecture. Brain, 121, 1603–30.
Roehrenbach, C. and Landis, T. 1995: Dreamjourneys: Living in woven realities, the
syndrome of reduplicative paramnesia. In R. Campbell and M.A. Conway, (eds)
Broken Memories: Case Studies in Memory Impairment. Oxford: Blackwell, 93–100.
Romme, M.A. and Escher, A.D. 1989: Hearing voices. Schizophrenia Bulletin, 15,
209–16.
Seligman, M.E.P., Abramson, L.Y., Semmel, A. and von Baeyer, C. 1979: Depressive
attributional style. Journal of Abnormal Psychology, 88, 242–7.
Sharp, H.M., Fear, C.F. and Healy, D. 1997: Attributional style and delusions: an
investigation based on delusional content. European Psychiatry, 12, 1–7.
Sims, A. 1988: Symptoms in the Mind. London: Baillière Tindall.
Sno, H.N. 1994: A continuum of misidentification symptoms. Psychopathology, 27,
144–7.
Stone, T. and Young. A. 1997: Delusions and brain injury: The philosophy and psy-
chology of belief. Mind and Language, 12, 327–64.
Sweeney, P.D., Anderson, K. and Bailey, S. 1986: Attributional style in depression:
A meta-analytic review. Journal of Personality and Social Psychology, 50, 974–91.
Tranel, D., Damasio, H. and Damasio, A.R. 1995: Double dissociation between overt
and covert face recognition. Journal of Cognitive Neuroscience, 7, 425–32.
Weiner, B. 1988: Attribution theory and attributional therapy: some clinical obser-
vations and suggestions. British Journal of Clinical Psychology, 27, 93–104.
Weiner, B. 1990: Searching for the roots of applied attribution theory. In S. Graham
and V.S. Folkes (eds), Attribution Theory: Applications to Achievement, Mental Health,
and Interpersonal Conflict. Hillsdale, NJ: Lawrence Erlbaum Associates, 1–16.
Wright, S., Young, A.W. and Hellawell, D.J. 1993: Sequential Cotard and Capgras
delusions. British Journal of Clinical Psychology, 32, 345–9.
Young, A.W. 1994: Recognition and reality. In E.M.R. Crichley (ed.), The Neurologi-
cal Boundaries of Reality. London: Farrand Press, 83–100.
Young, A.W., Leafhead, K.M. and Szulecka, T.K. 1994: The Capgras and Cotard
delusions. Psychopathology, 27, 226–31.
Young, A.W., Reid, I., Wright, S. and Hellawell, D.J. 1993: Face-processing impair-
ments and the Capgras delusion. British Journal of Psychiatry, 162, 695–8.
Young, A.W., Robertson, I.H., Hellawell, D.J., de Pauw, K.W. and Pentland, B.
1992: Cotard delusion after brain injury. Psychological Medicine, 67, 361–70.

The Cognitive Neuropsychology of Delusions

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Cognitive Neuropsychology of Delusions

Uploaded by

Copyright:

Available Formats

The Cognitive Neuropsychology of Delusions

ROBYN LANGDON AND MAX COLTHEART

A false belief based on incorrect inference about external reality that

Mind & Language, Vol. 15 No. 1 February 2000, pp. 184–218.

despite what constitutes incontrovertible and obvious proof or evi-

Despite concern about how such a definition should be operationalized by

The second is to evaluate models of normal cognition in terms of how well

2. The Role of Perceptual Aberrations

we can be aware that an intended response was wrong after we have

If we accept such a model of normal action control which involves two

whilst the insightful patient with depersonalization disorder describes his or

One of the historical forerunners of attribution theory is the ‘naı̈ve psychology’

that Capgras patients often present with a pervasive mood of

nalizing bias for negative events—cannot be jointly sufficient to explain the

butional bias which favours personal-level causal explanations over subper-

However, the presence of a perceptual aberration, although necessary to

5. Studies of Reasoning in Deluded Individuals

to non-deluded non-schizophrenic psychiatric controls and normal controls,

6. Cognitive Accounts of the Presence of Delusional Beliefs

Although there is much to be said for bias accounts of delusion formation

(1) first-person information which we access directly via our senses;

On this view, normal rational evaluation of the plausibility and probability of

7. A Cognitive Model of Normal Belief Generation and

cant (personally relevant) stimuli is controlled by a mechanism which differs

 Blackwell Publishers Ltd. 2000

8. Bizarre Two-Deficit Delusions

8.1 Capgras and Cotard Delusion

8.2 Alien Control Delusions

8.3 Auditory Hallucinations and Loss of Boundary Delusions

a source-monitoring defect coupled with the loss of subjective privacy nor-

8.4 Delusions of Reference

8.6 Non-Hallucinatory Polythematic Delusions

for example, loss of boundary experiences and delusions of reference. Now,

9. The Implications of a Single Deficit of Normal Belief

 Blackwell Publishers Ltd. 2000

10. Summary and Conclusions

Macquarie Centre for Cognitive Science

 Blackwell Publishers Ltd. 2000

Graham, S. and Folkes, V.S. 1990: Attribution theory: Applications to achievement,

 Blackwell Publishers Ltd. 2000

You might also like